Articles, and Fungal Hyphae Also Physically Derson Et Al., 2008)

Home , Chernozem

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ doi:10.5194/soil-1-83-2015 SOIL © Author(s) 2015. CC Attribution 3.0 License.

An ecosystem approach to assess soil quality in organically and conventionally managed farms in Iceland and Austria

J. P. van Leeuwen1, T. Lehtinen2,3,4, G. J. Lair3,5, J. Bloem6, L. Hemerik1, K. V. Ragnarsdóttir4, G. Gísladóttir2, J. S. Newton7, and P. C. de Ruiter1 1Biometris, Wageningen University, P.O. Box 100, 6700 AC Wageningen, the Netherlands 2Institute of Life and Environmental Sciences, University of Iceland, Sturlugata 7, IS-101 Reykjavík, Iceland 3Institute of Soil Research, University of Natural Resources and Life Sciences (BOKU), Peter-Jordan-Straße 82, 1190 Vienna, Austria 4Institute of Earth Sciences, University of Iceland, Sturlugata 7, IS-101 Reykjavík, Iceland 5Institute of Ecology, University of Innsbruck, Sternwartestrasse 15, 6020 Innsbruck, Austria 6Alterra, Wageningen University and Research Centre, P.O. Box 47, 6700 AA Wageningen, the Netherlands 7Department of Biological Sciences, University of Alberta, CW 405, Edmonton T6G 2E9, Alberta, Canada Correspondence to: J. P. van Leeuwen ([email protected])

Received: 12 June 2014 – Published in SOIL Discuss.: 24 June 2014 Revised: – – Accepted: 17 October 2014 – Published: 6 January 2015

Abstract. Intensive agricultural production can be an important driver for the loss of long-term soil quality. For this reason, the European Critical Zone Observatory (CZO) network adopted four pairs of agricultural CZO sites that differ in their management: conventional or organic. The CZO sites include two pairs of grassland farms in Iceland and two pairs of arable farms in Austria. Conventional fields differed from the organic fields in the use of artificial fertilisers and pesticides. Soils of these eight farms were analysed in terms of their physical, chemical, and biological properties, including soil aggregate size distribution, soil organic matter contents, abundance of soil microbes and soil fauna, and taxonomic diversity of soil microarthropods. In Icelandic grasslands, organically farmed soils had larger mean weight diameters of soil aggregates than the conventional farms, while there were no differences on the Austrian farms. Organic farming did not systematically influence organic matter contents or composition, nor soil carbon and nitrogen contents. Also, soil food web structures, in terms of presence of trophic groups of soil organisms, were highly similar among all farms, indicating a low sensitivity of trophic structure to land use or climate. However, soil organism biomass, especially of bacteria and nematodes, was consistently higher on organic farms than on conventional farms. Within the microarthropods, taxonomic diversity was systematically higher in the organic farms compared to the conventional farms. This difference was found across countries and farm, crop, and soil types. The results do not show systematic differences in physical and chemical properties between organic and conventional farms, but confirm that organic farming can enhance soil biomass and that microarthropod diversity is a sensitive and consistent indicator for land management.

Published by Copernicus Publications on behalf of the European Geosciences Union. 84 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

1 Introduction soil organisms will become more important in terms of deliv- ering important soil ecosystem functions, especially in terms Soil is considered to be one of the most important natural of soil structure formation, soil carbon dynamics, and nutri- resources for life on Earth. Soil processes govern a wide ar- ent mineralisation, as well as the suppression of soil-borne ray of ecosystem services, such as the provision of food, feed diseases (Birkhofer et al., 2008). The present study investi- and fibre, carbon sequestration, hydrological regulation, and gates biological, physical, and chemical soil quality parame- contaminant attenuation (Costanza et al., 1997). ters, focused on soil structure formation, soil organic matter Mostly due to human activities, soil quality, here defined dynamics and nutrient cycling, and the soil as a habitat for in terms of the soil’s ability to deliver ecosystem services, species-rich communities. is being drastically reduced in many locations worldwide Soil structure is an important attribute of soil quality. Soil (Vitousek, 1997). Global loss of soil ecosystem services is aggregates and the pores between the aggregates provide due to many different environmental threats, such as climate space, water, and oxygen, thereby creating habitats for a large change, intensive agricultural production, and environmental diversity in soil organisms (Anderson, 1978; Sulkava and pollution. Huhta, 1998). Soil organisms play an important twofold role In order to come up with effective strategies to protect and in determining soil structure formation. Firstly, microorgan- enhance soil quality, the Critical Zone Observatory (CZO) isms produce exudates (polysaccharides) that enhance aggre- network was established across the USA and Europe (An- gation of soil particles, and fungal hyphae also physically derson et al., 2008). The CZO network is an internationally bind soil particles (de Gryze et al., 2005; Wright et al., 2007; coordinated interdisciplinary research effort to better under- Tisdall and Oades, 1982). Secondly, the soil fauna plays a stand the chemical, physical and biological processes that role in creating a stable soil pore structure through moving in shape the Earth’s surface and support the terrestrial life on the soil and the formation of faecal pellets (Oades, 1993; Lee the planet. and Foster, 1991; Jastrow and Miller, 1991; Lavelle et al., As part of the CZO research effort, the European Com- 2006). Furthermore, soil structure is strongly linked to soil mission has provided funding for a large multi-disciplinary organic matter (SOM) dynamics, as incorporation of SOM research project: Soil Transformations in European Catch- into the soil aggregates “protects” it from microbial decom- ments (SoilTrEC). This project aims to understand and quan- position, thereby stabilising SOM content and sequestering tify the physical, chemical, and biological processes that are carbon in the soil, with potentially positive effects on plant critical to soil ecosystem functions and services in the Euro- productivity (Golchin et al., 1994). pean CZOs (Bernasconi et al., 2011; Menon et al., 2014). Soil organic matter is an essential component of soil qual- The European CZO network consists of sites along soil ity, governing processes like carbon sequestration, nutrient formation gradients (Austria, Switzerland, Iceland), along a cycling, water retention, and soil aggregate turnover. Soil or- soil degradation gradient (Greece), along a pollution gradi- ganic matter dynamics are driven by land use through root ent (Czech Republic), and of agricultural sites differing in turnover, deposition of plant residues, and decomposition by methods of soil management (Austria, Iceland) (Menon et the soil microbial populations. Soil organisms are known to al., 2014; Banwart et al., 2011). play important roles in SOM dynamics (Wardle et al., 2004; This paper presents the soil quality assessment as carried de Ruiter et al., 1994; Lavelle et al., 2006) by decompos- out for the agricultural CZO sites in Europe. The agricul- ing SOM. This process mineralises carbon (C) and nutrients tural sites have been chosen as part of the CZO network be- like nitrogen (N), making these available for plant uptake. To cause intensive agricultural production is an important driver understand the role of soil organisms in decomposition pro- of loss in soil quality, e.g. due to decreased organic matter cesses, SOM has been defined in terms of fractions based on contents. Intensive agriculture may also cause environmen- decomposability (Golchin et al., 1994). The idea behind this tal problems, e.g. nitrate leaching to nearby natural ecosys- fractioning is that the labile fractions, such as dissolved and tems, and pesticide contamination of surface and groundwa- particulate organic matter, are better available for biological ter (Skinner et al., 1997). The agricultural CZO sites con- decomposition, contribute more to soil structure formation, sist, in total, of eight farms: four grassland farms in Ice- and are more sensitive to soil management than more sta- land, of which two are conventional and two organic, and ble fractions such as lignin (Beare et al., 1994; Tisdall and four arable farms in Austria, of which two are conventional Oades, 1982). and two organic. The organic farms differed from the con- The soil as habitat for species-rich communities has in- ventional farms in that only organic fertilisers were applied creasingly received attention for the intrinsic and functional and no pesticides were used. On the conventional grassland value of soil biodiversity. High levels of biodiversity are farms in Iceland, some organic fertilisation was used in addi- thought to enhance stability of soil functions and services tion to the artificial inorganic fertilisers. On the conventional against perturbations and disturbances, and aid in the sup- arable farms in Austria, only artificial inorganic fertilisers pression of soil-borne pests and diseases (Griffiths et al., were applied together with pesticides. The central idea be- 2000; Altieri, 1999; Barrios, 2007). Soil biodiversity is also hind the organic farming practice is that the community of recognised as a sensitive biological indicator for effects of

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 85 environmental change and disturbance (Wardle et al., 1995; 2.2 Sampling scheme Ritz et al., 2009; Pattison et al., 2008; Ponge et al., 2006). One of the key indicator groups is the soil microarthropods, Samples were taken in May–June 2011 (0–10 cm in Iceland, because these are abundant, functionally diverse, and re- 0–15 cm in Austria). On each farm, three plots were selected spond to a variety of ecological and environmental factors at which all measurements were carried out; the plots were (Gardi and Parisi, 2002; Parisi et al., 2005). In addition, the approximately 30–40 m apart. At each plot, mixed soil sam- area covered during the lifecycle is representative of the ex- ples (ca. 1 kg, from 10 to 15 cores) were taken by use of amined site, and their life histories permit insights into soil a 8 cm diameter corer for microbial (bacteria, fungi), mi- ecological conditions (Gardi et al., 2009). crofaunal (protozoa, nematodes), soil chemical and physi- The results presented in this paper are from a field sur- cal measurements, and a 5 cm diameter corer for the meso- vey on all agricultural CZO sites, in which soil was analysed fauna (enchytraeids and microarthropods). In the grasslands in terms of its physical, chemical, and biological properties. on Iceland, vegetation diversity was estimated by application Soil physical and chemical measurements included soil ag- of four 2 m line transects at all farms, except for the con- gregate size fractions ( < 20, 20–250, 250–5000 µm); soil or- ventional farm in southern Iceland, for which the vegetation ganic matter contents and distribution (based on different or- data were supplied by the farmer. A line-intercept method was applied and four 2 m length tapes were laid out from ganic matter fractions); nutrient contents, including nitrogen ◦ (N), phosphorus (P), and potassium (K); and soil pH. Soil bi- the sampling point, each tape separated by 90 . Species were ological measurements included the presence and abundance recorded each time a plant species intercepted the tape, or of soil microbes (bacteria, fungi) and soil fauna (protozoa, when a group of equally mixed plant species occurred (e.g. nematodes and microarthropods), representing the main tax- Kent and Coker, 1992). Vegetation richness was calculated onomic groups and trophic levels in the soil food web. In as the total number of plant species present on the transects. addition we measured the taxonomic richness and diversity within the group of microarthropods, as well as vegetation 2.3 Soil physicochemical measurements diversity. Particle size distribution (clay content) was determined with 2 Methods a combined sieve and pipette method after removal of organic matter with hydrogen peroxide and dispersion by recip- 2.1 Site description rocal shaking with sodium metaphosphate solution for 12 h (Burt, 1992). Soil pH was measured electrochemically (mi- The soils analysed were sampled from the eight agricultural croprocessor pH meter pH196 WTW, Weilheim, Germany) CZO research sites; of these, four are under sub-arctic (Ice- in H2O at a soil : solution ratio of 1 : 2.5 (Burt, 1992). Cal- land) and four under continental (Austria) climatic condi- cium (Ca) content was measured by flame atomic absorp- tions. The four farms in Iceland were grassland farms, and tion spectrophotometry (Perkin-Elmer 2100). Plant available the four in Austria were arable farms practicing crop ro- P and K were determined by calcium acetate–lactate (CAL) tations (Tables 1, 2). In each country, two farms applied extraction (ÖNORM L1087). “organic” practices and two farms applied “conventional” A three-step procedure was carried out to fractionate practices. The organic farms differed from the conventional soil aggregates and organic matter. Free particulate or- farms in that only organic fertilisers were applied and no ganic matter (fPOM, 20–5000 µm) was separated using pesticides. On the conventional grassland farms in Iceland, sodium polytungstate solution (density of 1.8 g cm−3). To some organic fertilisation was used in addition to the artifi- obtain particulate organic matter occluded in aggregates cial inorganic fertilisers. The organic fields in Iceland were (oPOM, 20–5000 µm), the heavy fraction of soil aggregates ploughed the first three consecutive years when grasslands (> 1.8 g cm−3) was treated by ultrasound (8 J mL−1), which were renewed to apply green manure, whereas conventional disrupted the macroaggregates and protected the microaggre- fields were ploughed only once. On the conventional arable gates (Lehtinen et al., 2014). With a subsequent density frac- farms in Austria only artificial inorganic fertilisers were ap- tionation step (sodium polytungstate solution, 1.8 g cm−3), plied together with pesticides. In Iceland, one pair of or- the oPOM floating on the suspension was obtained after ganic and conventional farms (in the southwest) was on His- centrifugation (10 min at 4350 rpm). POM fractions were tic Andosols; the other pair (southern Iceland) was on Haplic washed with deionised water until the electric conductivity (Brown) Andosols. In Austria, one pair of organic and con- dropped below 5 µS cm−1 (Steffens et al., 2009). The residue ventional farms grew potatoes as the current crop; the other of the density fractionation procedure – mineral particles and pair grew winter wheat. All Austrian farms were situated in organo-mineral associations – was sieved at 250 and 20 µm the Marchfeld, southeast of Vienna, on Haplic Chernozems. to obtain macroaggregates (250–5000 µm) and microaggre- Farm properties are listed in Tables 1 and 2. gates (20–250 and < 20 µm). All aggregate fractions were washed with deionised water until the electronic conductivity dropped below 5 µS cm−1; subsequently they were oven- www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 86 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

Table 1. Characteristics of the farms studied in Iceland (conventional farms IceHaAcon and IceHiAcon, organic farms IceHaAorg and IceHiAorg), including vegetation richness (values represent mean and standard deviation (between brackets)).

Country Iceland Iceland Iceland Iceland Type Conventional Organic Conventional Organic Farm IceHaAcon IceHaAorg IceHiAcon IceHiAorg Coordinates N 64◦02033.78, N 64◦0300.2, N 64◦20032.82, N 64◦20042.90, W 20◦12018.06 W 20◦10044.4 W 21◦34054.42 W 21◦36014.22 Average temperature (◦C)∗ 3.6 3.6 4.3 4.3 Average rainfall (mm)∗ 1120 1120 800 800 Soil type Haplic Andosol Haplic Andosol Histic Andosol Histic Andosol Land use type Grassland Grassland Grassland Grassland Last tillage 1995 2003 1998 1996 Conversion to organic – 1996 – 1994 Organic fertilisers 20 (spring) 35 (spring) 30 (spring) 30 (spring) – Manure (t ha−1) – Compost (t ha−1) 35 (fall) 10 (fall) – Cattle urine (t ha−1) 50 (spring) – Total N (kg N ha−1) 40 970 60 260 – Total C (t C ha−1) 0.8 8.6 1.2 3.2 Inorganic fertilisers 80 (spring) 300 (spring) – Total N (kg ha−1) – Total P (kg ha−1) 20 (spring) – Total K (kg ha−1) 20 (spring) Vegetation richness 4 7 (0) 4 (0) 8 (1.73)

∗ Icelandic Meteorological Office database (2012). dried at 100 ◦C and weighed. The weights of aggregates were mineralisation were measured by incubation of 200 g of ho- corrected for the sand content of the same size (for aggre- mogenised and sieved soil for 6 weeks at 20 ◦C (Bloem et al., gates 20–250, and > 250 µm) in order to exclude a sand par- 1994). Results of the first week (disturbance) were not used. ticle from being weighed as an aggregate (Six et al., 2000; N mineralisation was calculated from the increase in mineral Lehtinen et al., 2014). Mean weight diameter (MWD) of the N (nitrate and ammonium) between week 1 and week 6. Total sand-corrected aggregates was calculated according to Kem- concentrations of O2 and CO2 were measured weekly using per and Rosenau (1986) as the sum of the geometric means a gas chromatograph (Carlo Erba GC 6000) equipped with a of aggregate sizes multiplied by the respective fraction. hotwire detector (HWD 430) and helium as carrier gas, and Total carbon (TC) and nitrogen (TN) in bulk soil, aggre- weekly rates were calculated from that. Only bottles in which gates, and POM fractions were quantified by dry combus- O2 concentration dropped below 15 % within the 6-week pe- tion using an elemental analyser (Carlo Erba NA 1500 anal- riod were flushed and reset to environmental concentrations yser). For the analysis, 5 g of sieved (< 2 mm) soil without to prevent O2 limitation. For the statistical analyses, we took visible roots and litter was ground to a size < 63 µm for ho- the average of weekly rates over the 5-week period after the mogenisation and 1–1.5 mg soil was used for the analysis. first week. Total organic carbon (TOC) was calculated as the difference of total and inorganic C, measured as carbonate C by treating 0.5–2 g of fine-ground soil material with 10 % HCl acid and 2.4 Soil food web measurements quantifying the evolved CO . Hot-water-extractable carbon 2 The soils were analysed for the presence and abundances (HWC) was measured as the C present in solution after 16 h of the major taxonomic groups of soil organisms: bac- at 80 ◦C, while water-soluble carbon (WSC) was measured teria, fungi, protozoa, nematodes, enchytraeids, and mi- after 30 min at 20 ◦C (Ghani et al., 2003). Labile carbon was croarthropods. Within these taxonomic groups we defined defined as HWC, while recalcitrant carbon was determined “trophic groups” based on diet and life-history traits, fol- as the difference between TOC and labile carbon. Potentially lowing the method of Moore et al. (1988). Abundances were mineralisable nitrogen (PMN) was measured as the increase transformed into estimates of biomass based on body-size in NH during 1 week of anoxic incubation in slurry at 40 ◦C 4 information, and expressed in units of kilograms of carbon (Canali and Benedetti, 2006). Potential carbon and nitrogen per hectare for the 0–10 cm top soil layer.

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 87 0 . ) 09 1 0 − 50 ◦ (kg ha 3, E 16 . 15 0 14 ◦ Crop Fertiliser c ) 1 − 05.1 N 48 0 50 (t ha ◦ 55.6, E 16 0 13 ◦ . 1 − Crop Horse manure ) 1 − 9 N 48 . (kg ha 20 0 41 ◦ 3, E 16 . total nitrogen content: 200–400 kg N ha 09 0 c ; 17 1 ◦ − Crop Fertiliser b ) 1 − 5 N 48 . 24 0 41 total nitrogen content: 115–164 kg N ha ◦ b 9.5 9.5 9.5 9.5 7, E 16 525 525 525 525 . 08 0 17 ◦ a C) ◦ a Characteristics of the farms studied in Austria (organic farms AusPOTcon and AusWWorg, conventional farms AusPOTcon and AusWWcon). Crop rotation before 2001 was Zentralanstalt für Meteorologie und Geodynamik, 2014; CountryTypeFarmCoordinatesCrop rotation history N 48 Austria2011 Crop2010 Organic AusPOTorg2009200820072006200520042003 Biowaste Potato compost (t ha 2002 Soy bean2001 Soy bean Winter wheat Potato Soy bean Corn Austria AusPOTcon Winter wheat Conventional Potato Poppy Winter wheat 10 10 10 10 10 10 10 10 Potato Sugar beet Onion Winter wheat AusWWorg Organic Winter Austria wheat Potato N N 118, 120 P Sugar beet 46, K N 60 95, P 50, K 130 Winter wheat Sugar N beet 120 Winter wheat Winter Winter wheat wheat N Potato 120 Spring barley N N 120 120 20 Peas AusWWcon Winter wheat Conventional Sugar beet Austria Winter wheat Spring barley & Corn Clover potato Corn mix Winter wheat 20 Sugar beet N 138, P 21, K Clover 21 mix N 150, P 40, K Winter 40 wheat N 126 Winter wheat Corn Sun ﬂowers Durum wheat N 138, P 40, N K 128, 40 P 24, K Sugar 24 beet N 60 N 129, P 19, Corn K 19 N 137 N 133, P 30, K Winter 30 wheat N 130, P 25, K 25 Soil typeLand use typeLast tillageConversion to organic Crop rotation (potato) 1976 Chernozem 2010 Crop rotation (potato) Chernozem – 2010 Crop rotation (winter wheat) Crop rotation (winter wheat) Chernozem 1995 2010 Chernozem – 2010 Average temperature ( Average rainfall (mm) Table 2. similar to the crop rotation presented in the table. a www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 88 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

Bacterial biomass, fungal biomass, leucine incorporation, and protozoa were measured after a pre-incubation period S ◦ X of 2 weeks at 20 C. Bacterial numbers and cell volumes, H = − (pi × ln pi), and fungal hyphal lengths were measured in microscopic i=1 slides (Bloem and Vos, 2004). Bacterial cell numbers and in which p is the proportion of the total biomass (S), i.e. the volumes were determined using confocal laser scanning mi- i relative biomass, of species i. For the Pielou evenness index croscopy combined with an image analysis system. The (J ) we used the formula data were transformed into bacterial biomass, taking a specific carbon content of 3.20 × 10−13 g C µm−3 (Bloem et H J = , al., 1995). For the transformation of fungal hyphal lengths ln(N) to fungal biomass we described fungal volume as a cylin- in which H represents the Shannon diversity index and N the der with spherical ends (V = (π/4)W 2 (L − W/3), where total number of taxa present. V = volume in µm3, L = length in µm, and W = diameter in µm), with a mean hyphal diameter of 2.5 µm and a specific carbon content of 1.30 × 10−13 g C µm−3. Bacterial growth 2.5 Statistics activity was estimated by measuring incorporation rates of The data were from eight farms that differed in various ways: 14 [ C]leucine (Bloem and Bolhuis, 2006). climate, soil type, soil management, and crop. There were no Two trophic groups of protozoa (flagellates and amoebae) real replicates, as the triplicate measurements for all variables were measured using the most-probable-number method were from plots on the same farm. Hence, we performed a (Bloem et al., 1994). Numbers were converted to biomass nested two-way ANOVA with two factors: country (Iceland– assuming a spherical shape with diameters of 4.6 and 9.1 µm Austria) and farm management (organic–conventional), and for flagellates and amoebae, respectively, and a volume-to-C farm as a random nested factor. By taking country as a fac- −13 −3 conversion factor of 1 × 10 C µm (Bloem et al., 1994). tor, we separated the grassland (Iceland) from the cropland Soil nematodes were counted in 9 mL of soil solution ex- (Austria) farms. By including farm as a random nested fac- tracted by Oostenbrink elutriators from 100 g of soil. Num- tor, we accounted for the variation among farms. We tested bers per trophic group (bacterivore, fungivore, herbivore, the differences between soil types separately using a one-way omnivore, predaceous) were derived from species composi- ANOVA with soil as a factor. All data were log-transformed tion in the samples (Bongers, 1988). Nematode biomasses to obtain homogeneity of variances. Statistical analyses were were calculated using fresh weight data from Didden et carried out using SPSS (20.0.0) and R (2.15.2). al. (1994) and taking a moisture content of 75 % and a carbon content of 40 % (Didden et al., 1994). Enchytraeid numbers were obtained through a (wet) ex- 3 Results traction using Baermann funnels with increasing light and 3.1 Soil physicochemical measurements heat each 30 min after the start of the extraction over a total extraction time of 3 h. Enchytraeid numbers were converted Many physicochemical soil characteristics varied strongly into biomass C by measuring the average fresh weight and over farms, as a consequence of different soil types, soil man- taking a moisture content of 85 % and a carbon content of agement, and climatic conditions (countries) (Table 3). The 50 % of the dry weight (Didden et al., 1994). most pronounced differences were found between the soils Microarthropods were extracted from four soil cores of from the two different countries. Clay content was lowest in 196 mL per replicate, during a 1-week period with Tull- the Haplic Andosols in Iceland (p = 0.001). Soils in Austria gren funnels, and processed using the gel-based sub-sample were alkaline (pH 8) as a result of the much higher calcium methodology (Jagers op Akkerhuis et al., 2008). Total num- content of the Chernozems, whereas the Andosols in Iceland bers were recorded, while species composition was assessed had a lower pH (pH 5–6). Plant available nutrients (P, K) in sub-samples of 100 individuals following Jagers op Akker- were much higher on the farms in Austria than in Iceland, huis et al. (2008), and references therein. Microarthropod due to the strong nutrient retention in Andosols (p = 0.001 biomass C was calculated based on individual weights, mois- and p = 0.026, Table 3). ture contents, and C contents from Didden et al. (1994). For the MWD of soil aggregates, we found a difference Microarthropod diversity was quantified in three ways: by between farm management: on the organic farms in Ice- absolute number of taxa present, by the Shannon diversity in- land the MWD was more than twice as high as on the dex (H), and by the Pielou evenness index. For the Shannon conventional farms, although the difference was not statis- diversity index (H) we used the following formula: tically significant (p = 0.173). The opposite was found in Austria, although here the differences were relatively small (Table 3). Mean weight diameter was positively correlated to fungal (Pearson test, r = 0.739, p = 0.006) and bacterial biomass (Pearson test, r = 0.664, p = 0.019), whereas

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 89 value p value p value p carbon mineralisation h potential mineralisable nitrogen; g water-soluble carbon; f hot-water-extractable carbon; e total soil organic carbon; d occluded particulate organic matter; c free particulate organic matter; b 47 354 (7192)716.8 (123.1) 51 597 (6967)11.77 (21.23) 904.1 (76.05) 88 3173128 (21 026) (676) 36.79 (1.44) 1931 (564.9)58.46 (16.83) 78 723 (5452) 55.85 3439 (5.29) (554) 71.21 (5.71) 2135 22 093 (243.0) (799)282.6 (96.90) 152.45 5615 (62.55) (1333) 65.21 (8.28) 502.4 (110.6) 215.9 27 792 (52.40) (6113) 162.33 (18.22) 745.9 488.5 (280.7) – (58.10) 28 5476 004 (477) (968) 13.67 (10.49) 565.8 (34.08) 1010 (82.75) 12.46 25 (4.20) 654 (2503) 1990 (101) 702.5 (59.91) 89.46 (67.10) 21.37 (7.39) 0.893 2232 (171) 26.21 – (65.86) 0.696 42.70 0.010 (14.45) 90.98 (28.47) 2074 0.072 (279) 0.876 0.577 97.41 (10.82) 0.905 2093 (116) 0.022 – 0.680 0.839 0.032 0.782 0.624 0.020 – 0.968 33.12 (15.45)5.69 (1.43) 23.46 (1.31) 444.12 7.22 (142.39) (1.71) 358.52 (66.42) 72.67 (50.11) 2.20 (0.97) 29.74 (18.37) 2.205069 (0.59) (238.6) 2.01 (0.32) 5113 (353.9) 2.63 (0.22) 2654 2.06 (641.8) (0.37) 3.54 (0.49) 2157 (1601) 1.69 (0.35) 3263 (506.4) 0.867 2.32 (0.42) 4467 (282.3) 0.185 4412 (148.9) 0.595 0.865 4544 0.203 (261.5) 0.584 0.914 0.507 0.572 74.8 (7.29)3.75 (1.42)15.86 96.2 (14.88) (10.2) 3.43 (0.68) 28.04 (21.70) 129 (20.0) 7.97 (6.64) 8.10 (2.23) 190 (38.9) 6.02 (1.16) 4.79 (1.17) 37 317.86 868 (69.63) (6963) 180.77 (38.54) 109.30 (19.53) 42 176 164.88 (1102) (40.63) 161.01 (38.62) 124.17 110 (8.12) 542 (5014) 281.92 (24.77) 107 955 123.07 (10 (13.58) 926) 0.758 0.955 0.785 0.026 0.001 0.001 0.698 0.999 0.859 ) ) ) 0.79 (–) 0.74 (–) 0.46 (–) 0.63 (–) 1.45 (–) 1.54 (–) 1.40 (–) 1.38 (–) 1 1 3 − − − yr ) yr ) ) ) ) 1 ) 1 ) 1 1 1 1 1 1 1 − − − − − − − − − ) ) 1 ) 1 1 − Soil physicochemical properties and biologically mediated processes on the farms studied in Iceland (conventional farms IceHaAcon and IceHiAcon, organic farms IceHaAorg − (g kg − (mm) 8.27 (3.75) 19.91 (4.10) 4.75 (1.20) 11.70 (0.84) 9.98 (4.57) 7.64 (2.37) 4.50 (2.41) 3.82 (2.24) 0.236 0.169 0.125 (g kg (kg ha (kg ha O) 5.76 (0.22) 5.88 (0.17) 5.07 (0.13) 5.17 (0.17) 7.92 (0.04) 7.95 (0.02) 8.04 (0.02) 8.12 (0.03) 0.757 0.001 0.934 (kg ha (kg ha (kg ha (kg ha c a 2 b h i e f g d nitrogen mineralisation rate. i HWC WSC Total N (kg ha PMN CountryTypeFarm Iceland Conventional IcelandBulk IceHaAcon density (g Organic cm IceHaAorg Iceland Conventional IceHiAconTOC Organic IceHiAorg Iceland Conventional AusPOTcon Organic AustriaC min AusPOTorg AusWWcon Conventional Austria Organic AusWWorg Austria Farming Country Austria Interaction Effect Effect Effect MWD fPOM oPOM N min Clay content (%)pH (H Ca (kg ha 5.23 (1.28) 5.43 (1.12) 15.93 (10.04) 13.72 (2.66) 17.02 (0.94) 16.70 (1.62) 13.93 (0.24) 14.40 (1.76) 0.995 0.165 0.997 P (kg ha K (kg ha Aggregate size distribution (mean weight diameter); Table 3. and IceHiAorg) and Austria (conventionalbrackets) farms per farm, AusPOTcon measured and in AusWWcon,shown. the organic topsoil farms (0–10 AusPOTcon cm). Signiﬁcance and values AusWWorg). of Values represent the mean factors farming and (organic standard vs. deviation conventional), (between country (Iceland vs. Austria), and the interaction effect are a rate; www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 90 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

Figure 1. Soil food web diagram representative for all eight farms. Boxes represent the presence of trophic groups in the soil food web, and arrows represent feeding interactions based on diet information. Solid groups were present at all farms, and dashed groups were only present at some farms. no significant correlations were found with organic matter OC and N to bulk soil (61 and 69 % for OC and 56 and 62 % parameters. The content of free particulate organic matter for N, respectively). On the winter wheat farms in Austria, (fPOM) and occluded particulate organic matter (oPOM) microaggregates of 20–250 µm contributed the greatest quan- varied strongly between the different countries and between tities of OC and N to bulk soil (46 and 50 % for OC and 45 soil types within countries. The fPOM content in the Ice- and 45 % for N, respectively), while on the potato farms in landic Histic Andosols (358–444 g kg−1) was higher than Austria the microaggregates < 20 µm contributed the great- in the Icelandic Haplic Andosols (23–33 g kg−1) and all est quantities of OC and N to bulk soil (51 and 46 % for OC Austrian soils (2–3 g kg−1, p < 0.001). The oPOM content and 51 and 47 % for N, respectively). showed a similar pattern. The high contents of particulate organic matter in Iceland, especially in the Histic Andosols, 3.2 Soil food web measurements reflect the very high content of organic carbon (contents of TOC, HWC, and WSC) and nitrogen (both total N and Based on presence–absence data of the soil organisms, we PMN) in these soils: TOC (p = 0.010), HWC (p = 0.072), constructed soil food web diagrams for all farms (Fig. 1). total N (p = 0.020), and PMN (p = 0.022) were all higher These diagrams were very similar; despite differences in cli- in Iceland compared to Austria. The farms on Histic An- matic conditions, crop type, soil type, and soil management, dosols in Iceland had a lower C mineralisation rate (2157– most of the trophic groups were present on all farms. Some 2654 kg ha−1 yr−1), but a much higher potential N minerali- of the trophic groups were only present at some farms, in- sation rate (746–1010 kg ha−1 yr−1), than the farms on Hap- cluding predaceous nematodes, bacterivore mites, herbofun- lic Andosols in Iceland; these differences were even more givore mites, and Diplura (Fig. 1, Table 4). pronounced compared to the farms in Austria (p = 0.032). Trophic groups showed differences in abundances The way organic carbon (OC) and nitrogen (N) were dis- (Table 3) and species composition (see microarthropod di- tributed over aggregate sizes and organic matter fractions, versity). Bacterial biomass was consistently higher on or- was also different between farms. On the organic farm on ganic farms in both countries, although the differences were Haplic Andosol in Iceland, macroaggregates > 250 µm con- not statistically significant. Bacterial activity, measured as tributed the greatest quantities of OC and N to bulk soil (65 % the incorporation rate of [14C]leucine, did not differ sig- OC, 65 % for N). On both farms on Histic Andosols in Ice- nificantly between farms. Fungal biomass did not show a land the fPOM fraction contributed the largest quantities of consistent pattern over all farms, although fungal biomass tended to be lower on the farms on Histic Andosols. Protozoa

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 91 value p value p value p of the trophic and taxonomic groups in the soil food webs, bacterial activity, ) 1 − 10.27 (15.01) 45.44 (27.95) 126.5 (64.52) 133.8 (37.45) 163.5 (78.72) 90.41 (9.38) 101.8 (8.30) 152.7 (50.74) 0.509 0.294 0.476 0.75 (0.92) 0.18 (0.12) 1.31 (0.87) 0.18 (0.07) 0.15 (0.09) 0.07 (0.01) 0.27 (0.04) 0.33 (0.07) 0.161 0.239 0.290 − 0.57 (0.06) 0.79 (0.03) 0.56 (0.08) 0.75 (0.11) 0.66 (0.04) 0.77 (0.06) 0.76 (0.08) 0.79 (0.08) 0.008 0.049 0.069 ) 1 − ∗ ) 1 − J h 1 index 1.33 (0.11) 2.38 (0.09) 1.28 (0.37) 1.91 (0.29) 1.60 (0.30) 2.07 (0.33) 2.05 (0.37) 2.41 (0.29) 0.027 0.176 0.311 − Biological parameters on the farms studied in Iceland (conventional farms IceHaAcon and IceHiAcon, organic farms IceHaAorg and IceHiAorg) and Austria (conventional H Bacterial activity: leucine incorporation rate. CountryTypeFarmBacterivore nematodesFungivore nematodesHerbivore nematodesOmnivore nematodes IcelandPredaceous nematodesTotal nematode biomass ConventionalEnchytraeids IceHaAcon Iceland Organic 0.07 (0.03)Bacterivore mitesFungivore IceHaAorg 0.003 mites (0.003) IcelandHerbofungivore 0.11 mites Conventional (0.03) 0.08 (0.02) IceHiAconNematovore 0.014 0.13 mites (0.013) (0.03) Organic ndOmnivore mites 0.001 0.33 0.12 (0.001) (0.02) (0.01) IceHiAorg 0.13Predaceous Iceland (0.05) mites 0.023 (0.007) 0.13Total (0.08) Acari Conventional 0.19 (0.02) AusPOTcon 0.021 0.07 0.35 (0.019) (0.03) (0.03) OrganicFungivore collembolans 0.02 Austria 0.033 (0.02) AusPOTorg (0.028)Herbofungivore collembolans 0.15 (0.09) nd 0.12 0.21 0.79 (0.03) (0.01) (0.94) nd 0.055Herbivore (0.032) collembolans AusWWcon nd 0.02 Conventional (0.03)Total collembolans 0.030 0.20 0.001 (0.008) (0.09) (0.001) 0.09 Austria 0.44 Organic AusWWorg (0.03) (0.14) 0.13 0.589Diplura (0.07) 0.02 0.006 (0.02) (0.011) 0.033 0.13 (0.040) (0.03) 0.21 0.28 (0.02) (0.10)Total nd microarthropod 0.52 0.009 biomass 0.09 0.049 0.004 (0.78) 0.030 (0.009) (kg Farming (0.09) 0.06 (0.003) (0.027) C Austria (0.11) ha 0.058 (0.074) nd 0.134 0.13 (0.071) (0.04) Country nd 0.003 0.03 nd 0.51 (0.003) (0.02) (0.22) nd 0.262 0.33 (0.26) 0.01 Interaction 0.013 (0.02) (0.013) 0.01 0.001 0.058 (0.01) 0.02 (0.001) (0.041) (0.027) 0.411 0.88 Austria 0.15 0.23 (0.60) (0.02) 0.076 (0.05) (0.044) 0.001 0.305 (0.001) (0.186) 0.15 0.09 (0.21) 0.17 (0.11) (0.09) 0.80 (0.56) 0.029 0.348 0.034 (0.014) 0.009 0.112 (0.022) nd 0.43 (0.008) (0.057) 0.035 (0.31) 0.48 (0.10) 0.001 0.003 0.06 (0.002) (0.004) 0.008 nd (0.02) 0.023 (0.006) nd 0.025 Effect 0.101 0.403 0.052 (0.029) (0.036) (0.106) (0.085) 0.02 0.945 (0.01) 0.001 nd (0.001) 0.005 0.700 0.004 (0.002) 0.015 0.310 0.009 0.021 (0.004) (0.004) (0.015) 0.12 0.58 (0.05) Effect (0.85) 0.811 nd 0.037 0.07 0.014 (0.005) 0.010 (0.05) 0.039 (0.008) (0.008) (0.012) 0.169 0.606 0.914 nd 0.060 Effect 0.16 0.001 (0.027) 0.07 0.003 0.188 (0.08) 0.337 (0.001) nd (0.06) (0.005) (0.014) nd 0.03 nd (0.01) 0.357 0.002 0.010 0.335 0.710 0.606 (0.004) (0.012) 0.001 (0.001) 0.09 0.08 (0.06) (0.05) nd 0.012 0.16 (0.009) nd 0.005 (0.04) 0.393 (0.005) 0.274 0.307 0.08 0.03 0.002 (0.08) 0.002 (0.01) (0.003) (0.002) 0.04 0.177 (0.004) 0.192 0.01 nd nd (0.01) 0.434 0.007 (0.009) 0.416 0.03 0.21 (0.01) (0.02) 0.012 nd 0.569 0.538 0.316 nd 0.357 0.05 0.13 (0.02) 0.050 (0.06) 0.131 0.153 0.350 nd nd 0.20 0.037 0.023 (0.01) 0.750 0.896 nd 0.002 (0.001) 0.649 0.069 0.466 0.053 0.191 0.374 nd 0.868 0.562 0.369 0.374 0.732 0.275 0.374 nd 0.562 nd 0.002 (0.004) nd 0.374 0.374 0.374 Microarthropod taxa richness (no. taxa)Shannon Pielou evenness 10.33 (1.15) 20.67 (2.08) 10.33 (4.04) 12.67 (1.53) 12.00 (5.29) 14.67 (3.21) 15.33 (4.16) 21.00 (1.00) 0.122 0.449 0.707 BacteriaLeu (pmol g FungiAmoebaeFlagellates 27.70 (3.41) 38.00 (3.51) 0.63 (0.24) 16.93 (8.70) 0.62 (0.35) 17.89 (6.32) 16.61 1.03 (2.79) (0.37) 30.04 (9.62) 0.31 (0.06) 4.33 (1.82) 55.49 0.82 (14.14) (0.49) 0.21 (0.02) 68.48 (15.49) 8.67 (1.76) 4.03 (2.03) 53.80 1.85 (9.86) (1.43) 18.34 (2.54) 3.02 94.88 (1.28) (22.20) 0.53 (0.26) 19.54 0.062 (6.94) 1.63 (0.09) 0.49 21.14 (0.18) (9.32) 0.005 2.68 (1.26) 15.00 (3.02) 1.08 (0.30) 0.920 0.736 3.04 (2.47) 0.78 (0.12)No. taxa/trophic group 0.187 0.315 0.655 0.522 0.101 0.587 0.122 0.448 1.03 (0.12) 2.07 (0.21) 1.03 (0.40) 1.27 (0.15) 1.20 (0.53) 1.47 (0.32) 1.53 (0.42) 2.10 (0.10) 0.122 0.449 0.707 Table 4. farms AusPOTcon and AusWWcon, organic farms AusPOTcon and AusWWorg): biomasses (kg C ha ∗ and microarthropod diversity. Numbers represent meanfarming and (organic standard vs. deviation (between conventional), brackets), country measured (Iceland in vs. the Austria), topsoil and (0–10 the cm); nd: interaction not effect detected. are Signiﬁcance shown. values of the factors www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 92 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

(amoebae, flagellates) and enchytraeids showed no clear pat- 4 Discussion tern in biomass (Table 4). Nematode biomass was consistently higher on organic In this study we investigated soil quality parameters (physi- farms than in conventional farms, regarding all trophic cal, chemical, and biological) on the organically and conven- groups, although differences were only statistically signifi- tionally managed farms that are part of the European CZO cant for herbivorous nematodes (p = 0.035) and total nema- network. tode biomass (p = 0.015, Fig. 2b). Microarthropod abundance varied strongly from just over −2 −2 4.1 Soil aggregate formation, soil organic matter, and 12 000 m to over 200 000 m . We did not find system- soil nutrient cycling atic differences between country or management type. To- tal microarthropod biomass was much higher on the conven- Regarding soil structure formation and soil organic matter, tional farms in Iceland compared to all other farms (Fig. 2c). the different farming practices, organic versus conventional, Total Acari biomass was significantly higher on conventional did not reveal systematic differences in many physical and farms compared to organic farms (p = 0.023, Table 4). The chemical soil properties. The soil aggregate size distribu- higher biomass of omnivorous mites (p = 0.012) and, to tions were consistently higher on organic than on conven- a lesser extent, also the consistently higher Acari biomass tional farms in Iceland, but no differences were found in Aus- (p = 0.023) on conventional farms was fully accounted for tria. Other management practices such as tillage (Beare et al., by the high biomass of the astigmatid mite Tyrophagus sim- 1994) or crop rotation history may have obscured effects of ilis. T. similis accounted for 98.1 and 99.7 % of the total om- organic amendments. For example, the arable farms in Aus- nivorous mite biomass, and 59.8 and 69.7 % of the total mi- tria applied a crop rotation with a yearly tillage. As soil ag- croarthropod biomass in the conventional grasslands in Ice- gregates are sensitive to soil tillage (Beare et al., 1994, 1997; land, while this species was (nearly) absent at all other farms Six et al., 2000), it could be expected that the differences be- (Appendix A). In Iceland, collembolan biomass was higher tween organic and conventional arable farms are comparably on conventional farms compared to organic farms (Table 4). small. In contrast, the Icelandic grasslands had not been tilled for 8–16 years (Table 1). Also, the addition of higher quanti- 3.3 Microarthropod species identity and diversity ties of organic amendments was expected to have a positive effect through enhanced soil biological activity, in terms of In total, 82 taxa of microarthropods were found in our study aggregate-forming substances. However, the observed higher sites, with an overall larger diversity in Austria than Iceland. mean weight diameters on the organic farms on Iceland could All farms showed striking differences in the microarthropod not be linked to higher organic matter contents, e.g. in terms species composition: only 3 out of the 82 taxa were present of total carbon, or a difference in organic matter compo- on all farms (the mesostigmatid Arctoseius cetratus and the sition. However, mean weight diameter of aggregates was prostigmatids Eupodes sp. and Pygmephorus sp.). In Iceland, significantly correlated with fungal and bacterial biomass. 27 taxa were found that did not occur in Austria, and 37 Both bacteria and fungi produce soil-binding compounds like taxa were found only in Austria, while only 18 taxa were polysaccharides, which are important for production of rel- found in both countries. The number of taxa only occurring atively small aggregates (de Gryze et al., 2005; Wright et on organic farms amounted to a total of 33, either in Iceland al., 2007). Soil fungi are assumed to be especially more im- (14 taxa) or in Austria (18 taxa), while 1 taxon (Tyrophagus portant for the formation of larger soil aggregates through sp.) was found on organic farms in both Iceland and Austria. entanglement by hyphae (Tisdall and Oades, 1982). Moreover, 12 taxa were found only on conventional farms, Regarding the soil carbon and nitrogen, we also did not de- of which 5 were in Iceland and 7 in Austria. The organic tect systematic differences between organic and conventional wheat farm in Austria had a remarkably high microarthropod farming. C and N mineralisation rates as well as the mea- taxonomic richness, with 34 taxa present, of which 12 were sured C and N pools (TOC, HWC, total N, PMN; Table 3) unique to that farm. The conventional grasslands in Iceland were quite similar on organic and conventional farms. Fur- in particular had low taxonomic richness of only 18 taxa (Hi- thermore, bacterial activity was similar on organic and con- Acon) and 17 taxa (HaAcon). ventional farms. The present results partly confirm the results Organic farms had a significantly higher microarthropod reported from earlier studies (van Diepeningen et al., 2006; diversity measured according to all diversity measures; for Bloem et al., 2006). the Shannon index (p = 0.027, Fig. 3a) and the Pielou index In summary, C and N contents and dynamics between or- for evenness (p = 0.008, Fig. 3b), differences were statisti- ganic and conventional farms have been studied in three dif- cally significant; for taxonomic richness it was not statisti- ferent ways: factorial field experiments on a single farm, pair- cally significant (p = 0.122, Fig. 3c). wise comparisons of farms (as in our study), and comparisons across a larger number of farms (n = 10–20). In a factorial field experiment on an arable farm, the Lovinkhoeve in the Netherlands, Bloem et al. (1994) found a higher C and

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 93

Figure 2. Biomass in kg C ha−1 of microbes (bacteria+fungi) (a), nematodes (b), and microarthropods (c) on organic and conventional farms in Austria and Iceland. Bars are means ± standard deviation (n = 6), measured in the topsoil (0–10 cm). P values are the results of a nested univariate analysis of variance (ANOVA), with type (conventional (white bars) or organic (grey bars)) and country (Austria or Iceland) as factors.

Figure 3. Shannon diversity index on microarthropod taxa (a), Pielou evenness on microarthropod taxa (b), and absolute microarthropod taxa richness (c) on organic and conventional farms in Austria and Iceland. Bars are means ± standard deviation (n = 6), measured in the topsoil (0–10 cm). P values are the results of a nested univariate analysis of variance (ANOVA), with type (conventional (white bars) or organic (grey bars)) and country (Austria or Iceland) as factors.

N mineralisation in an integrated field compared to a con- N dynamics are quite clear, while the effects on C dynamics ventional field, probably as a result of organic amendments. are not consistent. Similarly, on a grassland farm in the Netherlands, a higher In an example of a pairwise comparison between or- N mineralisation and potentially mineralisable N has been ganic and conventional arable farms in the Netherlands, van measured when organic fertiliser was applied, while no dif- Diepeningen et al. (2006) observed lower nitrate levels on or- ference has been found in C mineralisation (van Eekeren et ganic farms, with no differences in total organic C, organic al., 2009). Also, Poudel et al. (2002) found a higher potential N, or total N. Conventional farms in that study also applied N mineralisation in organically managed crop rotation fields organic manure in addition to artificial fertilisers, which is than in conventional fields in California, but there the organic comparable to the grasslands in Iceland, where we also did fields also grew legumes between growing seasons, enhanc- not find differences in total organic C and total N. In an ex- ing N availability. In Switzerland, Birkhofer et al. (2008) ob- ample of a comparison across larger number of farms in the served a lower N mineralisation when only mineral fertiliser Netherlands (n = 10–20), Bloem et al. (2006) showed higher was used, while C mineralisation did not show differences C and N mineralisation rates in organic grasslands compared between the fields. Also, in this study, no differences were to conventional grasslands, but not in the comparison be- found between organic fields and fields that received both ar- tween organic and conventional arable farms. Thus, our study tificial fertilisers and organic manure, similar to the Icelandic confirms the notion that more factors are variable and dif- grasslands in the present study. Thus, in factorial experiments ferences between organic and conventional farming are less on a single farm, the effects of organic management on soil prominent when C and N dynamics are studied on a larger scale with more farms involved. www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 94 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

4.2 Soil food web structure We lack an explanation for this somewhat unexpected result. For example, it is contrary to the results from an earlier The trophic structure of the soil food webs showed a high study, showing higher abundances of Acari in organic grass- similarity; nearly all trophic groups were present on all lands compared to conventional grasslands in Wales (Yeates farms. This indicates that the trophic structure of the soil et al., 1997). The similar collembolan biomass on organic food webs was neither very sensitive to management nor to and conventional farms is in line with the results of Birkhofer climate, soil type, and farm type. Biomass of the different et al. (2008) in Switzerland, but in contrast with the results organisms, however, differed between farms. of Bardgett et al. (1993), who reported higher collembolan Microbial biomass, as the sum of bacteria and fungi, was biomass in organic fields. The two species of collembolans consistently higher on organic farms, although not statisti- that are by far the most abundant in the study of Bardgett cally significant. The higher microbial biomass, especially et al. (1993) were much less abundant (Onychiurus procam- bacterial biomass, is in line with previous studies that have patus) or even absent (Folsomia quadrioculata) in our data, compared organic and conventional farms (Bloem et al., which may explain the difference between the studies. 2006; Hole et al., 2005; Haubert et al., 2009; Mäder et al., 2002; Birkhofer et al., 2008; van Diepeningen et al., 2006; 4.3 Microarthropod diversity Gunapala and Scow, 1998). Other studies also have reported a higher microbial activity (Bloem et al., 2006; Hole et al., The most systematic difference we found in the compar- 2005), which we did not find in our study. We did not find ison between organic and conventional farming was the differences in fungal biomass, in contrast with some previ- higher microarthropod diversity on the organically managed ous results (Yeates et al., 1997; de Vries et al., 2006), but in farms. This difference was found across countries, farm types line with others (Shannon et al., 2002). These results might (grassland versus arable), and crop and soil type. This find- be due to the fact that added organic amendments in or- ing is in agreement with Doles et al. (2001) and Macfadyen ganic farming are generally easily degradable and therefore et al. (2009). enhance mainly bacterial biomass and activity (Hole et al., Factors known to enhance soil microarthropod diversity 2005). include plant litter diversity leading to a higher microhabi- We observed a significantly higher total nematode biomass tat and resource diversity (Hansen and Coleman, 1998) and on organic farms. Although a higher biomass was observed plant species identity (Wardle et al., 2005). In Iceland, or- for all trophic groups of nematodes, the difference was ganic grasslands had a higher plant diversity than conven- mostly accounted for by herbivorous nematodes. This is in tional grasslands, which supports the hypothesis that plant di- agreement with the higher nematode abundance that was versity enhances belowground microarthropod diversity. On found after addition of organic manure to wheat fields in the arable farms in Austria, where plant diversity does not Switzerland, where herbivorous nematodes were also the play a role, the application of artificial fertilisers may have re- dominant group (Birkhofer et al., 2008). It is also in agree- duced the microarthropod diversity (Siepel and Van de Bund, ment with the higher nematode abundance (although dom- 1988). inated by fungivores) found in organic grasslands in Wales Soil microarthropod diversity is described as a sensitive (Yeates et al., 1997). Hence, our results support the notion biological indicator for effects of environmental change and that nematodes are sensitive to farming type and that they disturbance on soil quality (Gardi and Parisi, 2002; Parisi et profit from the addition of organic amendments. al., 2005; Gardi et al., 2009). Our results confirm that the Microarthropod biomass measurements did not reveal sys- taxonomic diversity of the soil microarthropods was sensitive tematic differences between farm types, although total mi- to differences in farm type and management system. croarthropod biomass was highest on the conventional farms If we look at these findings in terms of the role of bio- within Iceland. We also did not find a difference between the diversity in ecosystem functioning, we see that the higher grassland farms in Iceland and the arable farms in Austria. microarthropod diversity on organic farms did not result in This is a bit unexpected because it is frequently observed that differences in the food web structure, nor did it yield higher microarthropod biomass is higher in grasslands compared to ecosystem services, such as soil fertility or C sequestration. arable farms, because ploughing decreases microarthropod This is in agreement with Setälä et al. (2005), who argue that biomass, which is more intense for root/tuber crops such as the functional importance of individual groups is rather high potato (Vreeken-Buijs et al., 1998). In our study, the organic at coarse (trophic group) level but low at species level, and grasslands in Iceland were, however, ploughed in the three that effects of species diversity on ecosystem functioning are consecutive years when the field was renewed, which, to- most likely found in studies with a very low species rich- gether with the colder climatic conditions, may explain why ness and therefore a low functional redundancy. Neverthe- biomass of microarthropods was not higher in the grasslands less, in our study microarthropod diversity was found to be than in the arable fields (Sjursen et al., 2005). a sensitive and consistent indicator for land management. At We found a statistically higher biomass of mites (Acari) present, determining microarthropod diversity is a relatively on the conventional farms compared to the organic farms. intensive activity, but when the current progresses in method-

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 95 ology lead to faster and cheaper analyses, such as barcoding extracted microarthropods, soil microarthropod diversity will become more cost-effective and an even more valuable indicator for soil quality.

4.4 Conclusions In this study we investigated soil biological, chemical, and physical parameters for soil quality on organically and conventionally managed farms. The chosen farms were part of the European Critical Zone Observatory network. Factors that vary across farms – such as climate, soil type, and farm type, and the limited number of replicates taken – have made it difﬁcult to ﬁnd clear patterns or draw general conclusions. On the other hand, we did observe that the organic farms showed higher biological parameters, in particular the diversity in soil microarthropod diversity, despite these limi- tations. Physical and chemical parameters showed no clear differences between the organic and conventional farms. Our results therefore do support the use of microarthropod diversity as a soil quality indicator, although physical and chemical soil properties are indispensable for a complete assessment and understanding of soil quality.

www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 96 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

Appendix A

Table A1. Biomass (kg C ha−1) of the microarthropod taxa in the soil food web on the farms studied in Iceland (conventional farms Ice- HaAcon and IceHiAcon, organic farms IceHaAorg and IceHiAorg) and Austria (conventional farms AusPOTcon and AusWWcon, organic farms AusPOTcon and AusWWorg). Trophic groups: omnivorous mites (Ommi), bacterivorous mites (Bami), fungivorous mites (Fumi), nematovorous mites (Nemi), predatory mites (Prmi), herbofungivorous mites (HFmi), herbofungivorous collembolans (HFco), fungivorous collembolans (Fuco), and diplurans (Dipl). Numbers represent mean and standard deviation (between brackets), measured in the topsoil (0–10 cm).

Country Iceland Iceland Iceland Iceland Austria Austria Austria Austria Type Conventional Organic Conventional Organic Conventional Organic Conventional Organic Farm IceHaAcon IceHaAorg IceHiAcon IceHiAorg AusPOTcon AusPOTorg AusWWcon AusWWorg Acari Astigmata Ommi 0.0063 (0.011) Acaridae Astigmata Ommi 0.0010 (0.0018) Histiostoma Bami 0.0002 0.0007 (0.0002) (0.0009) Rhizoglyphus Fumi 0.0314 (0.0395) Schwiebea Fumi 0.0205 (0.0347) Tyrophagus Ommi 0.0003 0.0020 (0.0005) (0.0017) Tyrophagus Ommi 0.5194 0.7801 0.0002 0.0003 similis (0.7805) (0.5551) (0.0004) (0.0005) Mesostigmata Nemi 0.0020 0.0094 0.0003 0.0001 0.0012 0.0006 Alliphis sicu- (0.0034) (0.0086) (0.0005) (0.0002) (0.0011) (0.0010) lus Arctoseius Prmi 0.0007 (0.0012) Arctoseius ce- Prmi 0.0320 0.0031 0.0207 0.0186 0.0032 0.0015 0.0067 0.0169 tratus (0.0553) (0.0054) (0.0256) (0.0130) (0.0056) (0.0015) (0.0117) (0.0118) Arrhopalites Prmi 0.0011 caecus (0.0020) Dendrolaelaps Prmi 0.0011 (0.0010) Dendrolaelaps Prmi 0.0101 rectus (0.0174) Dendrolaelaps Prmi 0.0034 samsinaki (0.0058) Dendrolaelaps Prmi 0.0026 zwoelferi (0.0045) Dinychus per- Ommi 0.0010 foratus (0.0018) Evimirus Nemi 0.0001 uropodinus (0.0002) Hypoaspis Prmi 0.0006 0.0043 (0.001) (0.0075) Hypoaspis ac- Prmi 0.0025 uleifer (0.0044) Lysigamasus Prmi 0.0063 0.0043 0.0043 0.0011 (0.0059) (0.0048) (0.074) (0.0018) Lysigamasus Prmi 0.0178 0.0047 0.0082 runciger (0.0263) (0.0042) (0.0142) Pachylaelaps Prmi 0.0011 0.0141 karawaiewi (0.0019) (0.0082) Pergamasus Prmi 0.0008 0.0021 0.0014 (0.0014) (0.0036) (0.0025) Pergamasus Prmi 0.0019 norvegicus (0.0034) Prozercon Nemi 0.0007 0.0004 (0.0008) (0.0007) Rhodacarellus Prmi 0.0006 0.0046 (0.001) (0.0041) Rhodacarellus Prmi 0.0045 0.0115 silesiacus (0.0078) (0.0014) Rhodacaridae Prmi 0.0011 (0.002) Uropoda Prmi 0.0074 (0.0029) Uropoda Prmi 0.001 orbicularis (0.0017) Veigaia Prmi 0.0011 nemorensis (0.002) Veigaia plani- Prmi 0.0013 cola (0.0022) Oribatida HFmi 0.0001 Liebstadia (0.0001) similis

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 97

Table A1. Continued.

Country Iceland Iceland Iceland Iceland Austria Austria Austria Austria Type Conventional Organic Conventional Organic Conventional Organic Conventional Organic Farm IceHaAcon IceHaAorg IceHiAcon IceHiAorg AusPOTcon AusPOTorg AusWWcon AusWWorg Liochthonius HFmi 0.0003 (0.0005) Liochthonius HFmi 0.0008 propinquus (0.0014) Microppia minus Fumi 0.0001 (0.0002) Oromurcia sudet- HFmi 0.0014 ica (0.0009) Pantelozetes Fumi 0.0005 0.0001 0.0002 paolii (0.0004) (0.0002) (0.0003) Platynothrus HFmi 0.0009 thori (0.0016) Protoribates Fumi 0.0008 capucinus (0.0008) Rhysotritia ardua HFmi 0.0003 0.0006 (0.0004) (0.0006) Tectocepheus ve- Ommi 0.001 0.0009 0.0046 0.0049 latus (0.0018) (0.0008) (0.004) (0.0043) Trhypochthonius Ommi 0.0084 cladonicola (0.0022) Prostigmata Ommi 0.0018 0.0018 0.0027 0.0102 0.0018 0.0008 0.0569 0.0081 Eupodes (0.0019) (0.0012) (0.0047) (0.0044) (0.0012) (0.0003) (0.0201) (0.0065) Microtydeus Ommi 0.0002 0.001 0.0013 0.0003 0.0088 0.0033 (0.0003) (0.0018) (0.0017) (0.0003) (0.0009 (0.0057) Nanorchestes Ommi 0.0539 0.0143 0.0909 0.0229 (0.0506) (0.0062) (0.0206) (0.0032) Pyemotes Prmi 0.0023 (0.0024) Pygmephorus Fumi 0.0001 0.001 0.0039 0.0025 0.0012 0.0006 0.0079 0.004 (0.0002) (0.001) (0.0034) (0.0028) (0.0014) (0.0005) (0.0075) (0.0021) Rhagidia Prmi 0.0028 0.0035 0.0021 (0.0025) (0.0049) (0.0036) Scutacarus Ommi 0.0016 0.0007 0.0002 (0.0015) (0.0012) (0.0003) Speleorchestes Ommi 0.0095 0.009 0.0037 0.0004 (0.0029) (0.0029) (0.0025) (0.0007) Stigmaeidae Prmi 0.0432 (0.0549) Tarsonemus Ommi 0.004 0.0016 (0.004) (0.0015) Trombidiidae Prmi 0.0013 (0.0022) Tydeidae Ommi 0.0083 0.0007 (0.0084) (0.0007) Collembola Entomobryomorpha HFco 0.0066 0.1007 0.0089 (0.0026) (0.1064) (0.0035) Folsomia sex- oculata Folsomides Fuco 0.0006 0.0015 parvulus (0.0011) (0.0026) Isotoma Fuco 0.0006 0.0048 (0.001) (0.0083) Isotoma angli- Fuco 0.0045 0.0009 cana (0.0078) (0.0015) Isotomiella minor Fuco 0.0006 0.0056 0.0416 0.0307 0.0032 0.0005 (0.0011) (0.0038) (0.0607) (0.0226) (0.0017) (0.0009) Lepidocyrtus HFco 0.0054 (0.0094) Lepidocyrtus cya- HFco 0.008 0.0006 0.0014 neus (0.0097) (0.0011) (0.0024) Parisotoma nota- Fuco 0.0371 0.0366 0.0068 0.0042 0.0219 0.0221 bilis (0.0643) (0.0434) (0.0118) (0.0046) (0.0022) (0.0093) Proisotoma min- Fuco 0.0364 0.0338 uta (0.0631) (0.0024) Pseudisotoma Fuco 0.0036 sensibilis (0.0062) Pseudosinella HFco 0.0012 0.0024 0.0051 alba (0.0021) (0.0041) (0.0054) Neelipleona HFco 0.0016 0.0012 0.0027 0.0051 Megalothorax (0.0017) (0.0021) (0.0047) (0.0054) minimus Poduromorpha Fuco 0.0952 0.0414 0.2088 0.0017 0.0105 0.0024 0.0959 Ceratophysella (0.0525) (0.0481) (0.092) (0.003) (0.0167) (0.0041) (0.0284) denticulata Friesea truncata Fuco 0.0006 0.0024 0.0077 (0.0011) (0.0041) (0.0073) Hypogastrura Fuco 0.01 (0.0054) Mesaphorura Fuco 0.0132 (0.0131)

www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 98 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

Table A1. Continued.

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 99

Acknowledgements. This research was funded by the EU FP7- Bloem, J., Lebbink, G., Zwart, K. B., Bouwman, L. A., Burg- ENV-2009 project SoilTrEC “Soil Transformations in European ers, S. L. G. E., de Vos, J. A., and de Ruiter, P. C.: Dynamics Catchments” (contract no. 244118). This research was also sup- of microorganisms, microbivores and nitrogen mineralization in ported by the research programme KB IV “Innovative scientific re- winter-wheat fields under conventional and integrated manage- search for sustainable green and blue environment” funded by the ment, Agr. Ecosyst. Environ., 51, 129–143, doi:10.1016/0167- Netherlands Ministry of Economic Affairs, Agriculture and Innova- 8809(94)90039-6, 1994. tion, and carried out by Wageningen University and Research Cen- Bloem, J. and Vos, A.: Fluorescent staining of microbes for total tre. direct counts, in: Molecular microbial ecology manual,edited by: We thank An Vos, Meint Veninga, Popko Bolhuis, and Tamas Kowalchuk, G. A., de Bruijn, F. J., Head, I. M., Akkermans, A. D. Salanki for technical assistance. Wim Dimmers and Gerard Jagers L., and van Elsas, J. D., Kluwer Academic Publishers, Dordrecht, are acknowledged for counting and identifying microarthropods. 2nd Edn., 861–874, 2004. We are grateful to the farmers for access to their land, and for Bloem, J., Veninga, M., and Shepherd, J.: Fully automatic determi- providing land management information. nation of soil bacterium numbers, cell volumes, and frequencies of dividing cells by confocal laser scanning microscopy and im- Edited by: S. Grandy age analysis, Appl. Environ. Microbiol., 61, 926–936, 1995. Bloem, J. and Bolhuis, P. R.: Thymidine and leucine incorporation to assess bacterial growth rate, in: Microbiological methods for assessing soil quality, edited by: Bloem, J., Hopkins, D. W., and References Benedetti, A., CABI Publishing, Wallingford, UK, Cambridge, MA, 142–149, 2006. Altieri, M. A.: The ecological role of biodiversity in agroecosys- Bloem, J., Schouten, A. J., Sørensen, S. J., Rutgers, M., van der tems, Agr. Ecosyst. Environ., 74, 19–31, 1999. Werf, A., and Breure, A. M.: Monitoring and evaluating soil Anderson, J.: Inter-and intra-habitat relationships between wood- quality, in: Microbiological methods for assessing soil quality, land cryptostigmata species diversity and the diversity of soil and edited by: Bloem, J., Hopkins, D. W., and Benedetti, A., CABI litter microhabitats, Oecologia, 32, 341–348, 1978. Publishing, Wallingford, UK, 23–49, 2006. Anderson, S. P., Bales, R. C., and Duffy, C. J.: Critical zone obser- Bongers, T.: De nematoden van Nederland: Een identificatietabel vatories: Building a network to advance interdisciplinary study voor de in Nederland aangetroffen zoetwater-en bodembewo- of earth surface processes, Mineral. Mag., 72, 7–10, 2008. nende nematoden, Koninklijke Nederlandse Natuurhistorische Banwart, S., Bernasconi, S. M., Bloem, J., Blum, W., Brandao, Vereiniging, 1988. M., Brantley, S., Chabaux, F., Duffy, C., Kram, P., Lair, G., Burt, R.: Soil survey laboratory methods manual, USDA, 1992. Lundin, L., Nikolaidis, N., Novak, M., Panagos, P., Ragnars- Canali, S. and Benedetti, A.: Soil nitrogen mineralization, in: dottir, K. V., Reynolds, B., Rousseva, S., de Ruiter, P., van Microbiological methods for assessing soil quality, edited by: Gaans, P., van Riemsdijk, W., White, T., and Zhang, B.: Soil Bloem, J., Hopkins, D. W., and Benedetti, A., CABI Publishing, processes and functions in critical zone observatories: Hypothe- Wallingford, UK, 127–135, 2006. ses and experimental design, Vadose Zone J., 10, 974–987, Costanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Han- doi:10.2136/vzj2010.0136, 2011. non, B., Limburg, K., Naeem, S., O’Neill, R. V., and Paruelo, J.: Bardgett, R. D., Frankland, J. C., and Whittaker, J. B.: The effects of The value of the world’s ecosystem services and natural capital, agricultural management on the soil biota of some upland grass- Nature, 387, 253–260, 1997. lands, Agr. Ecosyst. Environ., 45, 25–45, 1993. de Gryze, S., Six, J., Brits, C., and Merckx, R.: A quantifi- Barrios, E.: Soil biota, ecosystem services and cation of short-term macroaggregate dynamics: Influences of land productivity, Ecol. Econom., 64, 269–285, wheat residue input and texture, Soil Biol. Biochem., 37, 55–66, doi:10.1016/j.ecolecon.2007.03.004, 2007. doi:10.1016/j.soilbio.2004.07.024, 2005. Beare, M. H., Hendrix, P. F., and Coleman, D. C.: Water- de Ruiter, P. C., Neutel, A.-M., and Moore, J. C.: Modelling food stable aggregates and organic matter fractions in conventional- webs and nutrient cycling in agro-ecosystems, Trends Ecol. Evo- and no-tillage soils, Soil Sci. Soc. Am. J., 58, 777–786, lut., 9, 378–383, 1994. doi:10.2136/sssaj1994.03615995005800030020x, 1994. de Vries, F. T., Hoffland, E., van Eekeren, N., Brussaard, L., and Beare, M. H., Hu, S., Coleman, D., and Hendrix, P.: Influences of Bloem, J.: Fungal/bacterial ratios in grasslands with contrast- mycelial fungi on soil aggregation and organic matter storage in ing nitrogen management, Soil Biol. Biochem., 38, 2092–2103, conventional and no-tillage soils, Appl. Soil Ecol., 5, 211–219, 2006. 1997. Didden, W. A. M., Marinissen, J. C. Y., Vreekenbuijs, M. J., Burg- Bernasconi, S. M., Bauder, A., Bourdon, B., Brunner, I., Bünemann, ers, S., de Fluiter, R., Geurs, M., and Brussaard, L.: Soil meso- E., Chris, I., Derungs, N., Edwards, P., Farinotti, D., and Frey, fauna and macrofauna in 2 agricultural systems – factors affect- B.: Chemical and biological gradients along the damma glacier ing population-dynamics and evaluation of their role in carbon soil chronosequence, Switzerland, Vadose Zone J., 10, 867–883, and nitrogen dynamics, Agr. Ecosyst. Environ., 51, 171–186, 2011. 1994. Birkhofer, K., Bezemer, M. T., Bloem, J., Bonkowski, M., Chris- Doles, J. L., Zimmerman, R. J., and Moore, J. C.: Soil microarthro- tensen, S., Dubois, D., Ekelund, F., Fließbach, A., Gunst, L., pod community structure and dynamics in organic and conven- and Hedlund, K.: Long-term organic farming fosters below and tionally managed apple orchards in western Colorado, USA, aboveground biota: Implications for soil quality, biological con- Appl. Soil Ecol., 18, 83–96, 2001. trol and productivity, Soil Biol. Biochem., 40, 2297–2308, 2008. www.soil-journal.net/1/83/2015/ SOIL, 1, 83–101, 2015 100 J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality

Gardi, C. and Parisi, V.: Use of microarthropods as biological indi- ity in different soil orders quantified by low dispersive ul- cators of soil quality: The bsq sinthetic indicator, 7. International trasonic energy levels, Soil Sci. Soc. Am. J., 78, 713–723, Meeting on: Soils with Mediterranean Type of Climate, Valen- doi:10.2136/sssaj2013.02.0073, 2014. zano (Italy), 23–28 September 2001, 2002. Macfadyen, S., Gibson, R., Polaszek, A., Morris, R. J., Craze, P. G., Gardi, C., Montanarella, L., Arrouays, D., Bispo, A., Lemanceau, Planqué, R., Symondson, W. O. C., and Memmott, J.: Do differ- P., Jolivet, C., Mulder, C., Ranjard, L., Römbke, J., and Rutgers, ences in food web structure between organic and conventional M.: Soil biodiversity monitoring in Europe: Ongoing activities farms affect the ecosystem service of pest control?, Ecol. Lett., and challenges, Euro. J. Soil Sci., 60, 807–819, 2009. 12, 229–238, 2009. Ghani, A., Dexter, M., and Perrott, K. W.: Hot-water extractable Mäder, P., Fliessbach, A., Dubois, D., Gunst, L., Fried, P., and Nig- carbon in soils: A sensitive measurement for determining impacts gli, U.: Soil fertility and biodiversity in organic farming, Science, of fertilisation, grazing and cultivation, Soil Biol. Biochem., 35, 296, 1694–1697, 2002. 1231–1243, doi:10.1016/s0038-0717(03)00186-x, 2003. Menon, M., Rousseva, S., Nikolaidis, N. P., van Gaans, P., Pana- Golchin, A., Oades, J., Skjemstad, J., and Clarke, P.: Soil structure gos, P., de Souza, D. M., Ragnarsdottir, K. V., Lair, G. J., Weng, and carbon cycling, Soil Research, 32, 1043–1068, 1994. L., and Bloem, J.: Soiltrec: A global initiative on critical zone re- Griffiths, B. S., Ritz, K., Bardgett, R. D., Cook, R., Christensen, search and integration, Environ. Sci. Pollut. Res., 21, 3191–3195, S., Ekelund, F., Sørensen, S. J., Bååth, E., Bloem, J., and de 2014. Ruiter, P. C.: Ecosystem response of pasture soil communities to Moore, J. C., Walter, D. E., and Hunt, H. W.: Arthropod regula- fumigation-induced microbial diversity reductions: An examina- tion of micro-and mesobiota in below-ground detrital food webs, tion of the biodiversity–ecosystem function relationship, Oikos, Ann. Rev. Entomol., 33, 419–435, 1988. 90, 279–294, 2000. Oades, J. M.: The role of biology in the formation, stabiliza- Gunapala, N. and Scow, K. M.: Dynamics of soil microbial tion and degradation of soil structure, Geoderma, 56, 377–400, biomass and activity in conventional and organic farming sys- doi:10.1016/0016-7061(93)90123-3, 1993. tems, Soil Biol. Biochem., 30, 805–816, doi:10.1016/S0038- Parisi, V., Menta, C., Gardi, C., Jacomini, C., and Mozzanica, E.: 0717(97)00162-4, 1998. Microarthropod communities as a tool to assess soil quality and Hansen, R. A. and Coleman, D. C.: Litter complexity and composi- biodiversity: A new approach in Italy, Agr. Ecosyst. Environ., tion are determinants of the diversity and species composition of 105, 323–333, 2005. oribatid mites (acari: Oribatida) in litterbags, Appl. Soil Ecol., 9, Pattison, A. B., Moody, P. W., Badcock, K. A., Smith, L. J., 17–23, 1998. Armour, J. A., Rasiah, V., Cobon, J. A., Gulino, L. M., and Haubert, D., Birkhofer, K., Fliessbach, A., Gehre, M., Scheu, S., Mayer, R.: Development of key soil health indicators for the and Ruess, L.: Trophic structure and major trophic links in con- Australian banana industry, Appl. Soil Ecol., 40, 155–164, ventional versus organic farming systems as indicated by carbon doi:10.1016/j.apsoil.2008.04.002, 2008. stable isotope ratios of fatty acids, Oikos, 118, 1579–1589, 2009. Ponge, J.-F., Dubs, F., Gillet, S., Sousa, J. P., and Lavelle, Hole, D. G., Perkins, A. J., Wilson, J. D., Alexander, I. H., Grice, P. P.: Decreased biodiversity in soil springtail communities: V., and Evans, A. D.: Does organic farming benefit biodiversity?, The importance of dispersal and landuse history in hetero- Biol. Conserv., 122, 113–130, 2005. geneous landscapes, Soil Biol. Biochem., 38, 1158–1161, Jagers op Akkerhuis, G. A. J. M., Dimmers, W. J., van Vliet, P. doi:10.1016/j.soilbio.2005.09.004, 2006. C. J., Goedhart, P. W., Martakis, G. F. P., and de Goede, R. G. Poudel, D. D., Horwath, W. R., Lanini, W. T., Temple, S. R., and M.: Evaluating the use of gel-based sub-sampling for assessing Van Bruggen, A. H. C.: Comparison of soil N availability and responses of terrestrial microarthropods (collembola and acari) leaching potential, crop yields and weeds in organic, low-input to different slurry applications and organic matter contents, Appl. and conventional farming systems in northern California, Agr. Soil Ecol., 38, 239–248, 2008. Ecosyst. Environ., 90, 125–137, 2002. Jastrow, J. D. and Miller, R. M.: Methods for assessing the effects Ritz, K., Black, H. I., Campbell, C. D., Harris, J. A., and Wood, of biota on soil structure, Agr. Ecosyst. Environ., 34, 279–303, C.: Selecting biological indicators for monitoring soils: A frame- doi:10.1016/0167-8809(91)90115-E, 1991. work for balancing scientific and technical opinion to assist pol- Kemper, W. D. and Rosenau, R. C.: Aggregate stability and size icy development, Ecol. Indic., 9, 1212–1221, 2009. distribution, in: Methods of Soil Analysis. Part 1. Physical and Setälä, H., Berg, M. P., and Jones, T. H.: Trophic structure and func- Mineralogical Methods, edited by: Klute, A., Agronomy Mono- tional redundancy in soil communities, Biological diversity and graph 9, 2nd Edn., ASA/SSSA, Madison, WI, USA, 425–442, function in soils, 236–249, 2005. 1986. Shannon, D., Sen, A., and Johnson, D.: A comparative study of the Kent, M. and Coker, P.: Vegetation description and analysis: A prac- microbiology of soils managed under organic and conventional tical approach, London, UK, John Wiley and Sons, X, 1992. regimes, Soil Use Manage., 18, 274–283, 2002. Lavelle, P., Decaëns, T., Aubert, M., Barot, S., Blouin, M., Bureau, Siepel, H. and Van de Bund, C.: The influence of management prac- F., Margerie, P., Mora, P., and Rossi, J. P.: Soil invertebrates tises on the microarthropod community of grassland, Pedobiolo- and ecosystem services, Euro. J. Soil Biol., 42, Supplement 1, gia, 31, 339–354, 1988. S3–S15, doi:10.1016/j.ejsobi.2006.10.002, 2006. Six, J., Paustian, K., Elliott, E., and Combrink, C.: Soil structure Lee, K. E. and Foster, R. C.: Soil fauna and soil structure, Soil Re- and organic matter i. Distribution of aggregate-size classes and search, 29, 745–775, 1991. aggregate-associated carbon, Soil Sci. Soc. Am. J., 64, 681–689, Lehtinen, T., Lair, G. J., Mentler, A., Gísladóttir, G., Ragnars- 2000. dóttir, K. V., and Blum, W. E. H.: Soil aggregate stabil-

SOIL, 1, 83–101, 2015 www.soil-journal.net/1/83/2015/ J. P. van Leeuwen et al.: An ecosystem approach to assess soil quality 101

Sjursen, H., Michelsen, A., and Holmstrup, M.: Effects of Vitousek, P. M.: Human domination of earth’s ecosystems, Science, freeze–thaw cycles on microarthropods and nutrient avail- 277, 494–499, 1997. ability in a sub-arctic soil, Appl. Soil Ecol., 28, 79–93, Vreeken-Buijs, M. J., Hassink, J., and Brussaard, L.: Relationships doi:10.1016/j.apsoil.2004.06.003, 2005. of soil microarthropod biomass with organic matter and pore size Skinner, J., Lewis, K., Bardon, K., Tucker, P., Catt, J., and Cham- distribution in soils under different land use, Soil Biol. Biochem., bers, B.: An overview of the environmental impact of agriculture 30, 97–106, 1998. in the uk, J. Environ. Manage., 50, 111–128, 1997. Wardle, D. A., Begon, M., and Fitter, A. H.: Impacts of disturbance Steffens, M., Kölbl, A., and Kögel-Knabner, I.: Alteration of soil on detritus food webs in agro-ecosystems of contrasting tillage organic matter pools and aggregation in semi-arid steppe topsoils and weed management practices, in: Advances in ecological re- as driven by organic matter input, Euro. J. Soil Sci., 60, 198–212, search, Academic Press, 105–185, 1995. 2009. Wardle, D. A., Bardgett, R. D., Klironomos, J. N., Setala, H., van Sulkava, P. and Huhta, V.: Habitat patchiness affects decomposition der Putten, W. H., and Wall, D. H.: Ecological linkages between and faunal diversity: A microcosm experiment on forest ﬂoor, aboveground and belowground biota, Science, 304, 1629–1633, Oecologia, 116, 390–396, 1998. 2004. Tisdall, J. and Oades, J. M.: Organic matter and water-stable aggre- Wardle, D. A., Bardgett, R. D., Usher, M. B., and Hopkins, D. W.: gates in soils, J. Soil Sci., 33, 141–163, 1982. How plant communities inﬂuence decomposer communities, Bi- van Diepeningen, A. D., de Vos, O. J., Korthals, G. W., and van ological diversity and function in soils, 119–138, 2005. Bruggen, A. H.: Effects of organic versus conventional manage- Wright, S. F., Green, V. S., and Cavigelli, M. A.: Glomalin in aggre- ment on chemical and biological parameters in agricultural soils, gate size classes from three different farming systems, Soil Till. Appl. Soil Ecol., 31, 120–135, 2006. Res., 94, 546–549, doi:10.1016/j.still.2006.08.003, 2007. van Eekeren, N., de Boer, H., Bloem, J., Schouten, T., Rutgers, M., Yeates, G. W., Bardgett, R. D., Cook, R., Hobbs, P. J., Bowling, de Goede, R., and Brussaard, L.: Soil biological quality of grass- P. J., and Potter, J. F.: Faunal and microbial diversity in three land fertilized with adjusted cattle manure slurries in comparison welsh grassland soils under conventional and organic manage- with organic and inorganic fertilizers, Biol. Fert. Soils, 45, 595– ment regimes, J. Appl. Ecol., 34, 453–470, 1997. 608, 2009.